CN108444924A - A method of differentiating tealeaves storage period using hyper-spectral image technique - Google Patents
A method of differentiating tealeaves storage period using hyper-spectral image technique Download PDFInfo
- Publication number
- CN108444924A CN108444924A CN201810211952.6A CN201810211952A CN108444924A CN 108444924 A CN108444924 A CN 108444924A CN 201810211952 A CN201810211952 A CN 201810211952A CN 108444924 A CN108444924 A CN 108444924A
- Authority
- CN
- China
- Prior art keywords
- sample
- tea
- image
- tealeaves
- storage period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003860 storage Methods 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 235000013616 tea Nutrition 0.000 claims description 128
- 244000269722 Thea sinensis Species 0.000 claims description 108
- 238000001228 spectrum Methods 0.000 claims description 39
- 238000007689 inspection Methods 0.000 claims description 16
- 230000029087 digestion Effects 0.000 claims description 8
- 230000003595 spectral effect Effects 0.000 claims description 8
- 238000011282 treatment Methods 0.000 claims description 8
- 239000000284 extract Substances 0.000 claims description 7
- 230000013011 mating Effects 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 238000000513 principal component analysis Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 239000005030 aluminium foil Substances 0.000 claims description 4
- 230000014759 maintenance of location Effects 0.000 claims description 4
- 238000002360 preparation method Methods 0.000 claims description 4
- 238000004611 spectroscopical analysis Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 235000006468 Thea sinensis Nutrition 0.000 claims description 2
- 235000020279 black tea Nutrition 0.000 claims description 2
- 238000005303 weighing Methods 0.000 claims description 2
- 230000004069 differentiation Effects 0.000 abstract description 10
- 238000004451 qualitative analysis Methods 0.000 abstract description 4
- 238000009659 non-destructive testing Methods 0.000 abstract description 2
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 7
- 238000001514 detection method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 235000000173 Camellia sinensis var sinensis Nutrition 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000002203 pretreatment Methods 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 240000007524 Camellia sinensis var. sinensis Species 0.000 description 1
- 244000241257 Cucumis melo Species 0.000 description 1
- 235000015510 Cucumis melo subsp melo Nutrition 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009614 chemical analysis method Methods 0.000 description 1
- 238000000701 chemical imaging Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000012850 discrimination method Methods 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 235000021393 food security Nutrition 0.000 description 1
- 239000003205 fragrance Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 239000002932 luster Substances 0.000 description 1
- 238000001819 mass spectrum Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 235000014347 soups Nutrition 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 239000013077 target material Substances 0.000 description 1
- 210000002105 tongue Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a kind of methods differentiating tealeaves storage period using hyper-spectral image technique, this method application hyper-spectral image technique combination supporting vector machine algorithm establishes qualitative analysis model, realize that the quick and precisely differentiation of tealeaves storage period, this method have the characteristics that fast analyze speed, efficient, at low cost, test favorable reproducibility, sample without pretreatment and be convenient for online non-destructive testing.
Description
Technical field
The present invention relates to the method for discrimination of tealeaves storage period a kind of more particularly to it is a kind of apply hyper-spectral image technique
Differentiate the method for tealeaves storage period.
Background technology
The cultivated area and tea yield in China tea place rank first in the world, and tealeaves is the traditional large outlet production in China
Product.But with the variation of supply and demand situation in international tea trade, China's tea export amount and volume of trade growth trend are returned
It falls, while the demand of domestic market is also bordering on saturation, it is unsalable tealeaves occur far beyond the market demand for the production of tealeaves
Phenomenon.During drinking tea, the quality of newly picked and processed tea leaves is better than old tea, this is because with the increase of storage time, the appearance of tealeaves
Irreversible variation all has occurred in quality and taste substance.Newly picked and processed tea leaves quality is good, also exactly old when price is high, and newly picked and processed tea leaves list
When tea is quit listing.In order to make profit, part retailer pretends to be newly picked and processed tea leaves using old tea, or old tea and newly picked and processed tea leaves doping are sold.These with
The secondary bad competitive method substituted the bad for the good, not only affects market order, compromises tealeaves brand, the coloring agent etc. that work in-process uses
Additive is even more to have threatened food security.During choosing, consumer is typically only by sense organ（See color and luster, hear fragrance,
It samples tea soup）Newly picked and processed tea leaves and old tea are judged, is difficult to realize and effectively differentiates.Lack related side due to being directed to the detection of tealeaves storage period
Method and standard, therefore, a kind of not only science of research, accurate but also simple and efficient tealeaves storage period method of discrimination just seem particularly urgent
It cuts.
Tea leaf quality is judged at present mainly carries out sensory evaluation in terms of shape and endoplasm two, and assessment result is easily by tea-taster
Subjective factor influence.And the chemical analysis methods such as gas chromatography-mass spectrum is used in conjunction, electronic nose, electronic tongues, instrument price are high
It is expensive, operating method is complicated, can not realize tealeaves storage period rapid and convenient differentiate.Therefore, spectral technique and image analysis skill
The nondestructiving detecting means such as art gradually attract attention, but single detection means is extremely limited to the ability of Tea Quality Evaluation.It is high
Spectrum picture technology（Hyperspectral imaging, HSI）Visible light, near infrared light region in electromagnetic spectrum, with big
The spectral band of amount subdivision is imaged target material, can extract image features and more spectral signatures simultaneously
Parameter, it is more acurrate, comprehensively characterize measured matter.As a kind of green, fast and efficiently non-destructive testing technology, high-spectrum
As technology is the fusion of spectral technique and image processing techniques, spectral analysis capabilities and image resolution ability are had both, while to tea
Leaf inside and outside quality is detected, and effectively improves the reliability and stability of tea leaf quality evaluation result, disclosure satisfy that science meter
The flavor evaluation requirement of quantization.In recent years, hyper-spectral image technique is widely used in agricultural production, in Tea Production
It is identified using tree plant cultivation management monitoring, tea grades division and teas is then focused primarily upon.By hyper-spectral image technique application
In the differentiation of tealeaves storage period, tea sampled images to be measured and model are called, there is simplicity, just with chemical detection method compared with
The advantages that prompt, accurate, can reach the target for scientifically and accurately differentiating tealeaves storage period, make up the vacancy of Related Research Domain,
Newly picked and processed tea leaves phenomenon is palmed off to the old tea of strike tea market, regulating market order has positive effect.
Invention content
Pretend to be newly picked and processed tea leaves using old tea in sales of tea for current Some Enterprises, and traditional discrimination method is not easy to tealeaves
The present situation that storage period is differentiated, the purpose of the present invention is to provide a kind of method of discrimination of tealeaves storage period, this method utilizes
Visible/near infrared bloom spectrometer carries out Image Acquisition to the tealeaves of different storage, is sentenced with the EO-1 hyperion for obtaining tealeaves storage period
Other model, the storage period of tealeaves can be determined using the model, and method of discrimination is simple, differentiated that result is accurate, can be carried out
Batch is identified.
The present invention is achieved by the following technical solutions：
A method of differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that include the following steps：
（1）It is aged the preparation of tea sample：The tealeaves purchased is sub-packed in aluminium foil bag and seals, be divided into after measuring initial aqueous rate
Two condition storages；A part of tea sample is stored in freezer, and as antistaled tea sample, digestion time is set as 1 year, 0,90,
180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 altogether
Sample；Another part tea sample is stored at normal temperatures, and as storage tea sample, digestion time is set as 1 year, 0,90,180,270,
360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 samples altogether；
（2）The selection of the acquisition and sample calibration set, inspection set of high spectrum image：Using based on Vis/NIR instrument
High spectrum image system and mating acquisition and analysis software, are acquired the high spectrum image of sample, it is seen that/near infrared camera
Scanning range is 370.38-1036.54 nm, and the sampling interval is 0.55 nm, and every spectrum has 1232 variables, taken the photograph to EO-1 hyperion
After being set as head time for exposure and conveying device, to step（1）In the tea sample of different storage obtained be scanned, adopt
Before collecting sample image, first scanning standard white correcting plate and black correction image, and image is corrected to sample image using black and white
It is corrected, obtains the high spectrum image of the different storage tea sample, select a sample as test specimens every two samples
Product, i.e. 50 samples are as inspection set in 150 samples, remaining 100 sample is as calibration set；
（3）Dimension-reduction treatment and characteristics extraction：Dimension-reduction treatment is carried out to original high-spectral data using principal component analysis, before being based on
The weight coefficient of three principal component images extracts to obtain five characteristic wavelengths, chooses 100 × 100 pixel regions of picture centre
Center is as interested region（Range of interest, ROI）, 5 characteristic wavelength image ROI regions are extracted and preserved
Spectroscopic data as spectral signature information, the characteristics of according to tealeaves image, be based on gray level co-occurrence matrixes（Gray-level co-
Occurrence matrix, GLCM）The texture for calculating the general image of tiling tealeaves, extracts the textural characteristics under characteristic wavelength
Variable is as texture feature information；
（4）Establish two models：Optimized parameter is selected, establishes be based on support vector machines respectively（Support vector
Machine, SVM）Storage tea sample and antistaled tea sample storage period discrimination model；
（5）The inspection of model：Use step（2）In inspection set test to calibration set model as unknown sample；
（6）Unknown storage period Tea Samples are taken, by step（2）Hyper-spectral data gathering is carried out, established SVM models is utilized, passes through
Unknown storage period Tea Samples high spectrum image is analyzed, can fast qualitative determine tealeaves storage period.
The step（2）When middle progress high spectrum image acquisition, the tealeaves for weighing 11.5-12.5g is equably laid in rule
It is carried out in the culture dish that lattice are 9 cm of Ф.
The step（2）Before middle progress hyperspectral image data acquisition, it is seen that/near infrared camera acquisition parameter is set as exposing
8.5 ms between light time, conveying device speed are 1.15 mm/s.
The step（3）Middle use principal component analysis to filter out five characteristic wavelength for 670.74 nm, 720.08 nm,
836.14 nm, 886.09 nm and 936.05 nm.
The step（4）In model established based on SVM algorithm, kernel function is Radial basis kernel function（Radial basis
Function, RBF）, optimal solution is calculated by choosing penalty factor appropriate and nuclear parameter.
Described differentiates that the method for tealeaves storage period is suitable for removing part black tea and compressed tea using hyper-spectral image technique
All kinds of tealeaves in addition.
The analysis model that tealeaves storage period method of discrimination based on hyper-spectral image technique is established is for differentiating that tealeaves is stored
The purposes of phase.
Beneficial effects of the present invention are：
Tealeaves storage period method of discrimination of the present invention directly uses sample to detect as former state, saves cumbersome pre-treatment, mathematical model is built
After vertical, only need to carry out Image Acquisition by instrument with sample to be tested can analyze tealeaves storage period, disclosure satisfy that tealeaves is online
The demand of detection.With without claim sample, without calculate, without sample pre-treatments, without any chemical reagents, no sample damage,
Stablize the feature that quick, environmentally protective, easily operated, accuracy rate is high, reproducible.
Description of the drawings
Fig. 1 show the EO-1 hyperion primary light spectrogram of different storage tea sample in the present invention（a）Lu`an Guapian.Tea antistaled tea（b）
Lu`an Guapian.Tea stores tea（c）Mount Huang Mao Feng antistaled teas（d）Mount Huang Mao Feng stores tea.
Fig. 2 show the weight coefficient figure of first three in the present invention principal component image.
Fig. 3 show in the present invention image of six peace mating plate antistaled tea samples under 5 characteristic wavelengths.
Fig. 4 show in the present invention image of six peace mating plate storage tea samples under 5 characteristic wavelengths.
Fig. 5 show in the present invention image of Mao Feng antistaled tea samples in Mount Huang under 5 characteristic wavelengths.
Fig. 6 show in the present invention image that Mount Huang Mao Feng under 5 characteristic wavelengths stores tea sample.
Specific implementation mode
It elaborates below to the embodiment of the present invention, the present embodiment is carried out lower based on the technical solution of the present invention
Implement, gives detailed embodiment and specific operating process, but protection scope of the present invention is not limited to following implementation
Example.
Embodiment 1
The qualitative analysis that the present embodiment application visible/near infrared high spectrum image analytical technology combination supporting vector machine algorithm is established
Model checking tealeaves storage period, include the following steps：
（1）It is aged the preparation of tea sample：The Lu`an Guapian.Tea tea sample purchased is sub-packed in aluminium foil bag and seals after measuring initial aqueous rate
Mouthful, it is divided into two condition storages；A part of tea sample is stored in freezer, and as antistaled tea sample, digestion time is set as 1 year,
0,90,180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain altogether
150 samples；Another part tea sample is stored at normal temperatures, and as storage tea sample, digestion time is set as 1 year, 0,90,
180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 altogether
Sample；The tealeaves of 12 ± 0.5g is weighed as a sample, tealeaves is equably laid in the culture dish that specification is 9 cm of Ф,
Carry out high spectrum image acquisition；
（2）The selection of the acquisition and sample calibration set, inspection set of high spectrum image：Using TaiWan, China, five bell Optical Co., Ltd is based on
The high spectrum image system of Vis/NIR instrument and mating acquisition and analysis software carry out the high spectrum image of sample
Acquisition, it is seen that/near infrared camera scanning range is 370.38-1036.54 nm, and the sampling interval is 0.55 nm, and every spectrum has
1232 variables, after being set to EO-1 hyperion camera time for exposure and conveying device, before hyperspectral image data acquisition,
The time for exposure of good EO-1 hyperion camera is predefined to ensure the clear of image；The speed of conveying device is determined to avoid figure
As size and spatial resolution are distorted, it is seen that the setting of/near infrared camera acquisition parameter is as follows：Time for exposure is 8.5 ms, conveying
Device speed is 1.15 mm/s；To step（1）In the tea sample of different storage obtained be scanned, before acquiring sample image,
First scanning standard white correcting plate and black correction image, and sample image is corrected using black and white correction image, it obtains
The high spectrum image of the different storage tea sample selects a sample as test sample, i.e. 150 samples every two samples
In 50 samples as inspection set, remaining 100 sample is as calibration set；
（3）Dimension-reduction treatment and characteristics extraction：Dimension-reduction treatment is carried out to original high-spectral data using principal component analysis, before being based on
The weight coefficient of three principal component images extracts to obtain five characteristic wavelengths, chooses 100 × 100 pixel regions of picture centre
Center is as ROI region, Lu`an Guapian.Tea antistaled tea sample and the spectrum picture such as Fig. 1 for storing tea sample（a）（b）, first three principal component
Weight coefficient result such as Fig. 2 of image, selection obtain 5 characteristic wavelengths：670.74,720.08,836.14,886.09 and
936.05 nm；Image such as Fig. 3 of Lu`an Guapian.Tea antistaled tea sample and storage tea sample under 5 characteristic wavelengths, shown in 4, is extracted and preserved
The spectroscopic data of this 5 characteristic wavelength image ROI regions is as spectral signature information, the characteristics of according to tealeaves image, is based on
GLCM calculates the texture of the general image of tiling tealeaves, and the textural characteristics variable extracted under characteristic wavelength is believed as textural characteristics
Breath；
（4）Establish model：Optimized parameter is selected, the Lu`an Guapian.Tea antistaled tea sample for introducing radial basis function and storage tea sample are established
Storage period SVM discrimination model, can obtain Lu`an Guapian.Tea antistaled tea sample and the storage best penalty factor c of tea sample be respectively 64,
111.431, RBF nuclear parameter g are respectively 0.047,0.009；Lu`an Guapian.Tea antistaled tea sample and storage tea sample calibration set differentiation rate point
Not Wei 100% and 99%, model calibration set differentiation rate is higher；
（5）The inspection of model：Use step（2）In inspection set test to calibration set model as unknown sample；Six peace melons
Piece antistaled tea sample and storage tea sample differentiation rate are respectively 100% and 96%, can realize the accurate differentiation of storage period.
Embodiment 2
The qualitative analysis that the present embodiment application visible/near infrared high spectrum image analytical technology combination supporting vector machine algorithm is established
Model checking tealeaves storage period, include the following steps：
（1）It is aged the preparation of tea sample：The Huangshan Maofeng tea sample purchased is sub-packed in aluminium foil bag and seals after measuring initial aqueous rate
Mouthful, it is divided into two condition storages；A part of tea sample is stored in freezer, and as antistaled tea sample, digestion time is set as 1 year,
0,90,180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain altogether
150 samples；Another part tea sample is stored at normal temperatures, and as storage tea sample, digestion time is set as 1 year, 0,90,
180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 altogether
Sample；The tealeaves of 12 ± 0.5g is weighed as a sample, tealeaves is equably laid in the culture dish that specification is 9 cm of Ф,
Carry out high spectrum image acquisition；
（2）The selection of the acquisition and sample calibration set, inspection set of high spectrum image：Using TaiWan, China, five bell Optical Co., Ltd is based on
The high spectrum image system of Vis/NIR instrument and mating acquisition and analysis software carry out the high spectrum image of sample
Acquisition, it is seen that/near infrared camera scanning range is 370.38-1036.54 nm, and the sampling interval is 0.55 nm, and every spectrum has
1232 variables；After being set to EO-1 hyperion camera time for exposure and conveying device, before hyperspectral image data acquisition,
The time for exposure of good EO-1 hyperion camera is predefined to ensure the clear of image；The speed of conveying device is determined to avoid figure
As size and spatial resolution are distorted, it is seen that the setting of/near infrared camera acquisition parameter is as follows：Time for exposure is 8.5 ms, conveying
Device speed is 1.15 mm/s；To step（1）In the tea sample of different storage obtained be scanned, before acquiring sample image,
First scanning standard white correcting plate and black correction image, and sample image is corrected using black and white correction image, it obtains
The high spectrum image of the different storage tea sample.Select a sample as test sample, i.e. 150 samples every two samples
In 50 samples as inspection set, remaining 100 sample is as calibration set；
（3）Dimension-reduction treatment and characteristics extraction：Dimension-reduction treatment is carried out to original high-spectral data using principal component analysis, before being based on
The weight coefficient of three principal component images extracts to obtain five characteristic wavelengths, chooses 100 × 100 pixel regions of picture centre
Center is as ROI region, Mount Huang Mao Feng antistaled teas sample and the spectrum picture such as figure such as Fig. 1 for storing tea sample（c）（d）, choose and obtain 5
A characteristic wavelength：670.74,720.08,836.14,886.09 and 936.05 nm；Mount Huang Mao Feng is fresh-keeping under 5 characteristic wavelengths
Tea sample and the image such as Fig. 5 for storing tea sample, shown in 6, are extracted and preserved the spectroscopic data of this 5 characteristic wavelength image ROI regions
As spectral signature information；The characteristics of according to tealeaves image, calculates the texture of the general image of tiling tealeaves, extraction based on GLCM
Textural characteristics variable under characteristic wavelength is as texture feature information；
（4）Establish model：Optimized parameter is selected, the Mount Huang Mao Feng antistaled teas sample for introducing radial basis function and storage tea sample are established
Storage period SVM discrimination model, can obtain Mount Huang Mao Feng antistaled teas sample and the storage best penalty factor c of tea sample be respectively 0.435,
588.134, RBF nuclear parameter g are respectively 0.25,0.009；Mount Huang Mao Feng antistaled teas sample and storage tea sample calibration set differentiation rate reach
To 100%, model calibration set differentiation rate is higher；
（5）The inspection of model：Use step（2）In inspection set test to calibration set model as unknown sample；Mount Huang hair
Peak antistaled tea sample and storage tea sample differentiation rate reach 100% and 98%, can realize the accurate differentiation of storage period.
Embodiment 3
The qualitative analysis that the present embodiment application visible/near infrared high spectrum image analytical technology combination supporting vector machine algorithm is established
Model, includes the following steps quick discrimination tealeaves storage period：
（1）Unknown tea sample pretreatment：Unknown sample tea sample is taken, weighs the tealeaves of 12 ± 0.5g as a sample, tealeaves is equal
It is laid in evenly in the culture dish that specification is 9 cm of Ф；
（2）The acquisition of high spectrum image：Use five height of the bell Optical Co., Ltd based on Vis/NIR instrument of instrument TaiWan, China
Spectrum picture system and mating acquisition and analysis software, are acquired the high spectrum image of unknown sample, it is seen that/near-infrared phase
Machine scanning range is 370.38-1036.54 nm, and the sampling interval is 0.55 nm, and every spectrum has 1232 variables；To EO-1 hyperion
After camera time for exposure and conveying device are set, before hyperspectral image data acquisition, predefines good EO-1 hyperion and take the photograph
As the time for exposure of head is to ensure the clear of image；The speed of conveying device is determined to avoid picture size and spatial resolution
Distortion.The setting of visible/near infrared camera acquisition parameter is as follows：Time for exposure is 8.5 ms, and conveying device speed is 1.15 mm/
s；
（3）Unknown sample measures：Established SVM models are called in Matlab softwares, and the spectrum of above-mentioned storage sample is believed
Breath and texture information are analyzed, can quick discrimination go out storage period of tealeaves.
Claims (6)
1. a kind of method differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that include the following steps：
（1）It is aged the preparation of tea sample：The tealeaves purchased is sub-packed in aluminium foil bag and seals, be divided into after measuring initial aqueous rate
Two condition storages；A part of tea sample is stored at normal temperatures, and as storage tea sample, digestion time is set as 1 year, 0,90,
180,270,360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 altogether
Sample；Another part tea sample is stored in freezer, and as antistaled tea sample, digestion time is set as 1 year, 0,90,180,270,
360 d are separately sampled, take five groups of tea samples, every group of tea sample that 30 samples, 30 g of each sample is taken to obtain 150 samples altogether；
（2）The selection of the acquisition and sample calibration set, inspection set of high spectrum image：Using based on Vis/NIR instrument
High spectrum image system and mating acquisition and analysis software, are acquired the high spectrum image of sample, it is seen that/near infrared camera
Scanning range is 370.38-1036.54 nm, and the sampling interval is 0.55 nm, and every spectrum has 1232 variables, taken the photograph to EO-1 hyperion
After being set as head time for exposure and conveying device, to step（1）In the tea sample of different storage obtained be scanned, adopt
Before collecting sample image, first scanning standard white correcting plate and black correction image, and image is corrected to sample image using black and white
It is corrected, obtains the high spectrum image of the different storage tea sample, select a sample as test specimens every two samples
Product, i.e. 50 samples are as inspection set in 150 samples, remaining 100 sample is as calibration set；
（3）Dimension-reduction treatment and characteristics extraction：Dimension-reduction treatment is carried out to original high-spectral data using principal component analysis, before being based on
The weight coefficient of three principal component images extracts to obtain five characteristic wavelengths, chooses 100 × 100 pixel regions of picture centre
The spectroscopic data of 5 characteristic wavelength image ROI regions is extracted and preserved as Spectral Properties reference as interested region in center
Breath the characteristics of according to tealeaves image, the texture of the general image of tiling tealeaves is calculated based on gray level co-occurrence matrixes, extracts characteristic wave
Textural characteristics variable under long is as texture feature information；
（4）Establish two models：Optimized parameter is selected, establishes the storage tea sample based on support vector machines and the storage of antistaled tea sample respectively
Tibetan phase discrimination model；
（5）The inspection of model：Use step（2）In inspection set test to calibration set model as unknown sample；
（6）Unknown storage period Tea Samples are taken, by step（2）Hyper-spectral data gathering is carried out, established SVM models is utilized, passes through
Unknown storage period Tea Samples high spectrum image is analyzed, can fast qualitative determine tealeaves storage period.
2. the method according to claim 1 for differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that institute
State step（2）When middle progress high spectrum image acquisition, the tealeaves for weighing 11.5-12.5g is equably laid in specification as 9 cm of Ф
Culture dish in carry out.
3. the method according to claim 1 for differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that institute
State step（2）Before middle progress hyperspectral image data acquisition, it is seen that/near infrared camera acquisition parameter is set as the time for exposure 8.5
Ms, conveying device speed are 1.15 mm/s.
4. the method according to claim 1 for differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that institute
State step（3）Middle use principal component analysis to filter out five characteristic wavelength for 670.74 nm, 720.08 nm, 836.14 nm,
886.09 nm and 936.05 nm.
5. the method according to claim 1 for differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that institute
State step（4）In model established based on SVM algorithm, kernel function is Radial basis kernel function, passes through and chooses penalty factor appropriate
Optimal solution is calculated with nuclear parameter.
6. the method according to claim 1 for differentiating tealeaves storage period using hyper-spectral image technique, which is characterized in that sentence
Other method is suitable for all kinds of tealeaves in addition to part black tea and compressed tea.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810211952.6A CN108444924B (en) | 2018-03-07 | 2018-03-07 | Method for discriminating storage period of tea by applying hyperspectral image technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810211952.6A CN108444924B (en) | 2018-03-07 | 2018-03-07 | Method for discriminating storage period of tea by applying hyperspectral image technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108444924A true CN108444924A (en) | 2018-08-24 |
| CN108444924B CN108444924B (en) | 2020-12-08 |
Family
ID=63195289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810211952.6A Active CN108444924B (en) | 2018-03-07 | 2018-03-07 | Method for discriminating storage period of tea by applying hyperspectral image technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108444924B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109470639A (en) * | 2018-11-07 | 2019-03-15 | 浙江大学 | Diaphania larva based on visible light and near-infrared high light spectrum image-forming and its method for quickly identifying to mulberry leaf damage |
| CN110084788A (en) * | 2019-04-15 | 2019-08-02 | 宁波天贝文化传播有限公司 | A kind of compressed tea appearance texture record and recognition methods |
| CN111524092A (en) * | 2020-03-16 | 2020-08-11 | 中国农业科学院茶叶研究所 | Nondestructive testing method for black tea withering degree index |
| CN112699756A (en) * | 2020-12-24 | 2021-04-23 | 中国农业科学院农业信息研究所 | Hyperspectral image-based tea origin identification method and system |
| CN113484309A (en) * | 2021-06-28 | 2021-10-08 | 安徽农业大学 | Method for monitoring black tea fermentation degree by hyperspectral coupling nano colorimetric sensor |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004061412A2 (en) * | 2002-12-23 | 2004-07-22 | Gentag, Inc. | Method and apparatus for wide area surveillance of a terrorist or personal threat |
| WO2006102415A3 (en) * | 2005-03-22 | 2009-04-16 | Applera Corp | Optical camera alignment |
| CN101692052A (en) * | 2009-08-31 | 2010-04-07 | 江苏大学 | Hyperspectrum image technique-based method and hyperspectrum image technique-based device for identifying true and false famous tea |
| US20110221926A1 (en) * | 2010-03-12 | 2011-09-15 | Kanaev Andrey V | Spectral Anomaly Detection in Deep Shadows |
| CN104897729A (en) * | 2015-06-04 | 2015-09-09 | 安徽农业大学 | Sorting method of storage time of melon slice tea by using electronic nose |
| CN107169407A (en) * | 2017-03-29 | 2017-09-15 | 广东工业大学 | Hyperspectral image classification method based on joint bilateral filtering and extreme learning machine |
-
2018
- 2018-03-07 CN CN201810211952.6A patent/CN108444924B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004061412A2 (en) * | 2002-12-23 | 2004-07-22 | Gentag, Inc. | Method and apparatus for wide area surveillance of a terrorist or personal threat |
| WO2006102415A3 (en) * | 2005-03-22 | 2009-04-16 | Applera Corp | Optical camera alignment |
| CN101692052A (en) * | 2009-08-31 | 2010-04-07 | 江苏大学 | Hyperspectrum image technique-based method and hyperspectrum image technique-based device for identifying true and false famous tea |
| US20110221926A1 (en) * | 2010-03-12 | 2011-09-15 | Kanaev Andrey V | Spectral Anomaly Detection in Deep Shadows |
| CN104897729A (en) * | 2015-06-04 | 2015-09-09 | 安徽农业大学 | Sorting method of storage time of melon slice tea by using electronic nose |
| CN107169407A (en) * | 2017-03-29 | 2017-09-15 | 广东工业大学 | Hyperspectral image classification method based on joint bilateral filtering and extreme learning machine |
Non-Patent Citations (4)
| Title |
|---|
| KUMAR: "Objects Detection in Hyperspectral Images Using Spectral Derivative", 《JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING》 * |
| 施彦 编著: "《神经网络涉及方法与实例分析》", 31 December 2009, 北京邮电大学出版社 * |
| 柳杨 编著: "《数字图像物体识别理论详解与实战》", 31 January 2018, 北京邮电大学出版社 * |
| 蒋帆: "基于高光谱分析技术的机炒龙井茶等级识别方法", 《农业工程学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109470639A (en) * | 2018-11-07 | 2019-03-15 | 浙江大学 | Diaphania larva based on visible light and near-infrared high light spectrum image-forming and its method for quickly identifying to mulberry leaf damage |
| CN110084788A (en) * | 2019-04-15 | 2019-08-02 | 宁波天贝文化传播有限公司 | A kind of compressed tea appearance texture record and recognition methods |
| CN110084788B (en) * | 2019-04-15 | 2022-12-20 | 虚境茶仓(宁波)网络科技有限公司 | Compressed tea appearance texture recording and identifying method |
| CN111524092A (en) * | 2020-03-16 | 2020-08-11 | 中国农业科学院茶叶研究所 | Nondestructive testing method for black tea withering degree index |
| CN111524092B (en) * | 2020-03-16 | 2024-03-12 | 中国农业科学院茶叶研究所 | Nondestructive testing method for black tea withering degree index |
| CN112699756A (en) * | 2020-12-24 | 2021-04-23 | 中国农业科学院农业信息研究所 | Hyperspectral image-based tea origin identification method and system |
| CN112699756B (en) * | 2020-12-24 | 2023-08-25 | 中国农业科学院农业信息研究所 | Hyperspectral image-based tea origin identification method and system |
| CN113484309A (en) * | 2021-06-28 | 2021-10-08 | 安徽农业大学 | Method for monitoring black tea fermentation degree by hyperspectral coupling nano colorimetric sensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108444924B (en) | 2020-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108444924A (en) | A method of differentiating tealeaves storage period using hyper-spectral image technique | |
| Lin et al. | Recent developments in vibrational spectroscopic techniques for tea quality and safety analyses | |
| Ren et al. | Using near-infrared hyperspectral imaging with multiple decision tree methods to delineate black tea quality | |
| CN110763698B (en) | Hyperspectral citrus leaf disease identification method based on characteristic wavelength | |
| CN105651713B (en) | A kind of green vegetables leaf chlorophyll quantitative detecting method based on computer image analysis | |
| CN103134767B (en) | Method for liquor quality identification through infrared spectrum revision | |
| CN105181912B (en) | A kind of Noninvasive Measuring Method of Freshness in rice storage | |
| Arendse et al. | Evaluation of biochemical markers associated with the development of husk scald and the use of diffuse reflectance NIR spectroscopy to predict husk scald in pomegranate fruit | |
| CN113484309B (en) | Method for monitoring black tea fermentation degree by hyperspectral coupling nano colorimetric sensor | |
| CN101692052B (en) | Hyperspectrum image technique-based method and hyperspectrum image technique-based device for identifying true and false famous tea | |
| CN105021562A (en) | Traceable method for rice origin and application thereof | |
| CN108956604A (en) | A method of Eriocheir sinensis quality is identified based on hyper-spectral image technique | |
| CN109031344A (en) | A kind of method of Full wave shape laser radar and high-spectral data joint inversion forest structural variable | |
| CN104977258B (en) | The quality detecting methods such as tealeaves/cosmetics based on two-dimensional correlation spectra | |
| CN104122225A (en) | Illegal tea additive identification method based on near-infrared spectrum technique | |
| Laddi et al. | Significant physical attributes affecting quality of Indian black (CTC) tea | |
| CN110108649A (en) | The fast non-destructive detection method of oil crops quality based on terahertz light spectral technology | |
| CN105181606B (en) | The method that cane sugar content is distributed in peanut is detected based on high light spectrum image-forming technology | |
| Li et al. | Identification of aged-rice adulteration based on near-infrared spectroscopy combined with partial least squares regression and characteristic wavelength variables | |
| CN110243805A (en) | Fishbone detection method based on Raman high light spectrum image-forming technology | |
| CN109470639A (en) | Diaphania larva based on visible light and near-infrared high light spectrum image-forming and its method for quickly identifying to mulberry leaf damage | |
| CN106940292A (en) | Bar denier wood raw material quick nondestructive discrimination method of damaging by worms based on multi-optical spectrum imaging technology | |
| CN110231302A (en) | A kind of method of the odd sub- seed crude fat content of quick measurement | |
| CN108444943A (en) | A kind of adulterated method for quick identification of coffee based on Fourier Transform Near Infrared | |
| KR20080075761A (en) | Development for discriminating geographical origin of green tea products from korea and japan by nir spectrophotometer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |